Papers › Convolutional Neural Network and Rule-Based Algorithms for Classifying 12-lead ECGs
Convolutional Neural Network and Rule-Based Algorithms for Classifying 12-lead ECGs
Bjørn-Jostein Singstad, Christian Tronstad
The objective of this study was to classify 27 cardiac abnormalities based on a data set of 43 101 ECG recordings. A hybrid model combining a rule-based algorithm with different deep learning architectures was developed. We compared two different Convolutional Neural Networks; a Fully Convolutional Neural Network and an Encoder Network, a combination of both, and with the addition of another neural network using age and gender as input. Two of these combinations were finally combined with a rule-based model using derived ECG features. The performance of the models was evaluated on validation data during model development using hold-out validation. Finally, the models were deployed to a Docker image, trained on the provided development data, and tested on the Challenge validation set. The model that performed best on the Challenge validation set was then deployed and tested on the full Challenge test set. The performance was evaluated based on a particular Challenge score. Our team, TeamUIO, achieved a Challenge validation score of 0.377, and a full test score of 0.206 for our best model. The score on the full test set placed us at 20th out of 41 teams in the official ranking.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Encoder | Accuracy(stratified10-fold) | 0,20±0,02 | #1 of 2 | Archive leaderboard | report |
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Encoder | F1(stratified10-fold) | 0,35±0,01 | #1 of 2 | Archive leaderboard | report |
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Encoder | F2(stratified10-fold) | 0,40±0,01 | #1 of 2 | Archive leaderboard | report |
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Encoder | G2(stratified10-fold) | 0,19±0,01 | #1 of 2 | Archive leaderboard | report |
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Encoder | PhysioNet Challenge score (test data) | 0.206 | #1 of 2 | Archive leaderboard | report |
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Encoder | PhysioNet Challenge score 2020 (validation data) | 0.377 | #1 of 2 | Archive leaderboard | report |
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Encoder | PhysioNet/CinC Challenge Score(stratified10-fold) | 0,37±0,03 | #1 of 2 | Archive leaderboard | report |
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Fully Convolutional Network | Accuracy(stratified10-fold) | 0,13±0,02 | #2 of 2 | Archive leaderboard | report |
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Fully Convolutional Network | F1(stratified10-fold) | 0,28±0,01 | #2 of 2 | Archive leaderboard | report |
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Fully Convolutional Network | F2(stratified10-fold) | 0,36±0,02 | #2 of 2 | Archive leaderboard | report |
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Fully Convolutional Network | G2(stratified10-fold) | 0,15±0,01 | #2 of 2 | Archive leaderboard | report |
| ECG Classification | PhysioNet Challenge 2020 | 1D CNN Fully Convolutional Network | PhysioNet/CinC Challenge Score(stratified10-fold) | 0,36±0,01 | #2 of 2 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Introduced by this paper: 1D CNN
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections